A Case-based Approach to the Feline Patient with vomiting and Increased Liver Enzymes (Part 1: Hepatic Lipidosis)

Transcription

1 A Case-based Approach to the Feline Patient with vomiting and Increased Liver Enzymes (Part 1: Hepatic Lipidosis) Joseph Taboada, DVM, Dipl. ACVIM (Internal Medicine) School of Veterinary Medicine, Louisiana State University Baton Rouge, Louisiana Most of the noninvasive techniques used in evaluating the feline patient with suspected hepatobiliary disease, such as hepatic enzymology, liver function tests, and hepatobiliary imaging are useful in localizing disease to the liver or biliary system, however, they can rarely be used to establish a definitive diagnosis or an accurate prognosis. Invasive techniques are usually required for this purpose. Invasive (laparotomy, laparoscopic, ultrasound guided percutaneous, or blind percutaneous) techniques are necessary to obtain liver tissue for cytologic or histopathologic evaluation. Techniques for biopsy of the liver may require anesthesia, special equipment, and carry the risk of complications. Fine needle aspiration (FNA) can be utilized to obtain liver tissue for cytology. FNA of the liver is a relatively safe procedure that is simple to perform while requiring no special equipment and minimal patient sedation. Together with the rest of the diagnostic evaluation, FNA can be a useful tool in establishing a diagnosis and prognosis in selected patients with hepatobiliary disease. Additionally, it is a non-invasive technique that can be useful in deciding which subset of patients may require biopsy. Hepatic Ultrasonography and Ultrasound Guided Biopsy Ultrasonography is becoming more and more routine in the evaluation of the patient with hepatobiliary disease, especially in the referral setting. Indeed, the availability of ultrasound is often used by the primary care practitioner when making the decision whether or not to refer a patient. Hepatobiliary ultrasonography is readily available and often considered the noninvasive procedure of choice for trying to differentiate between primary hepatic and post-hepatic causes of cholestasis, for identifying portosystemic vascular anomalies, and for identifying focal or multifocal hepatobiliary abnormalities. Recent work has tried to correlate hepatic ultrasonographic images with histopathologic findings. Hepatic lipidosis in cats and hepatocutaneous syndrome (superficial necrolytic dermatitis, necrolytic migratory erythema) in dogs are examples of diseases where findings on hepatic ultrasonography are helpful in making a diagnosis. Care must be taken not to over-interpret results, however, because unfortunately,

2 there can be tremendous variability in the echogenic pattern produced by a given disease. It is usually difficult, if not impossible, to predict the histopathologic infiltrate from the echogenic pattern observed. One of the advantages of ultrasonographic evaluation of the liver is that ultrasound may be used to guide placement of needles for biopsy or FNA. This technique is especially useful when investigating focal or multifocal disease that might be missed by blind techniques. Biopsy guides are available for most transducers to facilitate ultrasound-guided biopsy. Visualization of the needle and biopsy site during the procedure may improve accuracy of biopsy and detection of complications. Ultrasound guided biopsy is widely considered safer than other biopsy techniques but studies supporting or refuting this contention are lacking. The small gauge needle used for biopsy in many of the percutaneous ultrasound guided techniques can be a limiting factor when trying to interpret hepatic biopsies obtained. Numerous liver biopsy techniques have been described. Blind percutaneous techniques are the simplest and most cost effective in many situations. There is a perception, however, that they are less accurate than other techniques and carry a higher risk of complication. Few studies have looked at the relative risk and diagnostic accuracy of percutaneous biopsy in the dog and cat. A 0% to 8.4% risk of complication has been reported in the few small studies reported. There was positive correlation between biopsy and necropsy findings in 80% of the cases in one study. In man, larger studies (189,085 people biopsied) have identified a complication rate of 0.28% and a mortality rate of 0.03%. Anemia and cancer would appear to be factors positively correlating with an increased risk of complication. Type of needle used, operator experience, biopsy technique (transabdominal vs. transthoracic), and platelet count are not. Hepatic Fine Needle Aspirate (FNA) Fine needle aspirate of the liver is simple and requires no special equipment. It can be performed with a 6- or 12-cc syringe and a 22-gauge, 1.5- to 3.5-inch disposable hypodermic or spinal needle. The 22-gauge spinal needle is useful in larger animals because of the longer sizes available but is rarely necessary when being used in cats. The needle is inserted into the liver via a percutaneous transabdominal (in cats and most dogs) or transthoracic approach (in large deep chested dogs) and gentle suction (3-5 ml) is applied. While maintaining suction, the needle is gently but quickly thrust into the liver parenchyma and then brought back to the original position without exiting the liver. The suction is then released and the needle is

3 withdrawn. If the technique is properly performed, blood will rarely be noted in the syringe or needle hub as the entire specimen should remain in the needle. To transfer the specimen to a clean glass slide detach the needle, draw a few milliliters of air into the syringe, reattach the needle, and gently expel the liver sample onto the slide. Do not forcibly "blow" the specimen onto the slide as this may damage the cells and result in preparation artifacts. Squash prep smears or blood film techniques may be used to smear hepatic samples. The transabdominal technique can be performed either in dorsal or right lateral recumbency with the pelvis of the animal positioned slightly lower than the head. I prefer lateral recumbency because patient restraint is generally easier. With the animal in lateral recumbency the needle is inserted at the point where the left costal arch begins its dorsal ascent. It is angled at about 45 O the body wall. The left caudal thoracic mammary gland is usually just ve ntral and slightly caudal to the point where the needle is to be inserted. Once the peritoneal cavity has been entered the needle is brought parallel to the body wall and slowly advanced while gently feeling down for the liver with the tip of the needle. Once the liver is felt the needle is again angled at 45 O, the needle tip is placed into the liver parenchyma, and the aspirate is performed. FNA is most useful in evaluating patients with hepatomegaly but it often gives valuable information in patients with normal sized livers as well Diffuse infiltrative, inflammatory, and neoplastic diseases lend themselves best to an FNA diagnosis. FNA is less applicable to focal or multifocal diseases or diseases in which cells do not exfoliate easily, such as fibrosis or sarcoma. Kristensen, et al. recently described a classification scheme for interpretation of hepatic cytology: their categories include normal, hyperplastic, inflammatory, degenerative, necrotic, cholestatic, neoplastic, mixed reactions, other reactions, and non-diagnostic. Hepatic Cytology The predominant cell type in a normal hepatic FNA is the hepatocyte, often found in cohesive clusters or regular sheets. Hepatocytes are large polyhedral to rounded cells with abundant gray to basophilic cytoplasm. They have a single (occasionally two) eccentric nucleus with uniformly course chromatin and a small, distinct nucleolus. The cytoplasm is usually granular with a small amount of green bile pigment occasionally present. Small columnar epithelial cells of biliary origin may also be observed. Low numbers of macrophages (Kupffer's cells) with or without intracellular hemosiderin are sometimes seen. Because of the highly

4 vascular nature of the hepatic sinusoidal milieu a background of erythrocytes and blood-borne leukocytes is invariably present. Degenerative diseases are characterized by cytoplasmic vacuolar changes. The differential diagnosis for hepatic vacuoles includes fat, glycogen, hydropic degeneration, and storage diseases. Feline hepatic lipidosis is the characteristic example of diseases of this type. In the dog, glycogen deposition associated with steroid hepatopathy or hydropic degeneration associated with an ischemic or toxic insult is more likely. Extracellular deposition of amorphous material is seen in hepatic amyloidosis. It is not uncommon to see mild hepatocellular vacuolization in cats with a large variety of chronic diseases so care must be taken in interpreting the finding vacuolar hepatopathy in this species. Inflammatory specimens are characterized by increased numbers of inflammatory cells interspersed between normal and/or reactive hepatocytes. The predominant inflammatory cell type characterizes the inflammation present. A definitive diagnosis can be made in some protozoal or systemic fungal diseases based on the presence of identifiable organisms. Histoplasmosis is the systemic fungal disease most likely to be diagnosed via hepatic FNA. One of the most useful applications of hepatic FNA is the diagnosis of hepatic neoplasia. Lymphosarcoma, biliary carcinoma, or metastatic neoplasias are the most likely hepatic tumors to be diagnosed. Because FNA is usually a blind tissue sampling technique it is most applicable when the clinical evaluation suggests diffuse parenchymal disease. Focal diseases are less likely to be diagnosed by blind aspirate techniques. The accuracy of hepatic FNA can be improved by obtaining multiple aspirates so at least 3 aspirates taken from slightly different angles should be routinely obtained. Complications of hepatic FNA are extremely rare. Bleeding is rarely a clinically significant problem even in animals with coagulation abnormalities. Using a blind technique to aspirate cells from the liver will occasionally result in inadvertent gall bladder aspiration. Aspiration of the gall bladder rarely causes serious problems for the patient. In fact, with ultrasound guidance the technique is routinely used to sample bile in cases of suspected cholangitis or liver fluke infestation. It should be noted that while FNA is a quick and easy technique, the sample does not always accurately reflect the underlying histopathologic diagnosis. Few studies have looked at correlation between FNA cytology and histopathology in the dog and cat. There was a 66% correlation in one study. The fact that correlation is not 100% stresses the point that care should be taken when interpreting results that do not seem to fit the presenting clinical picture. Liver biopsy is still often needed for definitive diagnosis.

5 Feline Hepatic Lipidosis Feline hepatic lipidosis is the most common feline liver disease in many studies. It is characterized by massive hepatocellular accumulation of triglycerides accompanying a disruption in hepatic lipid metabolism that often results in severe liver dysfunction. Most cases are idiopathic but diabetes mellitus, prolonged starvation, over-nutrition, hyperthyroidism, and hyperparathyroidism are possible initiating causes. Female cats are affected almost twice as frequently as males. Chronic vomiting is the most common presenting sign. Anorexia, weight loss, icterus, and hypersalivation are also seen. Many affected cats are (or were) obese yet show significant muscle wasting at the time of diagnosis. Not all cats with hepatic lipidosis are obese. Total bilirubin, SAP, SALT, SAST, and GGT are usually increased (>2-5 times normal about 75% of the time). About half of affected cats will be hyperglycemic (glucose > 200 mg/dl). Abdominal radiographs may reveal mild hepatomegaly and ultrasound may reveal increased hepatic echogenicity. Diagnosis is dependent on demonstration of heavily vacuolated hepatocytes on fine needle aspirate or liver biopsy. Aggressive treatment is important. If an underlying disease process is evident it should be treated. General therapy should include treating dehydration, hypoglycemia (if present), hypokalemia, hypophosphatemia, and hepatoencephalopathy (lactulose 1-3 ml/cat adjust to maintain soft stool and metronidazole 7-10 mg/kg PO bid-tid). Vitamin K1 (0.5 to 1.5 mg/kg SQ) can be administered if cholestasis has resulted in a bleeding tendency. PIVKA will be increased if vitamin K1 is needed. Nutritional support is the most important aspect of therapy. Total caloric intake should be Kcal/kg/day. Protein supplementation is important. Diets based solely on carbohydrates may worsen the disease so moderate or even high protein diets such as Hills Prescription Diet p/d or a/d, Eukanuba Veterinary Diets Nutritional Recovery Formula /Canine & Feline, or Abbott Animal Health Clinicare Canine/Feline Liquid Diet or Clinicare RF should be used. Switch to a lower protein diet if signs of hepatoencephalopathy ensue. Dietary supplements that have been recommended but not critically evaluated in cats with hepatic lipidosis include l- carnitine ( mg/day), taurine ( mg/day), B-complex, zinc (7-10 mg/kg elemental Zn/day), and vitamin E ( mg/day). Force feeding or enteral feeding is invariably necessary to maintain appropriate caloric intake. Appetite stimulants may assist the owner who wishes to force feed their cat but will rarely result in enough of an increase in appetite to meet the nutritional needs appropriate to treatment goals. Cyproheptadine [Periactin ] 2 mg/cat, mirtazapine [Remeron ] 1/8 to ¼ of a 15 mg tablet per cat, and oxazepam [Serax ] 1 mg/kg

6 sid-bid may be used. Diazepam (0.1 ml IV) and Midazolam (2-5 mcg/kg IV) can also result in appetite stimulation. The effect of diazepam is usually short lived and causes significant sedation. Midazolam may cause a more lasting stimulation and less sedation. Care must be taken if benzodiazepines such as oxazepam or diazepam are used because they may worsen hepatoencephalopathy. Oral diazepam has been noted to occasionally be hepatotoxic. Enteral feeding will probably be needed in most cats with hepatic lipidosis. Esophagostomy, gastrostomy, or nasoesophageal feeding may all be used successfully. Gastrostomy and esophagostomy feeding is tolerated well by most cats. The tubes can be placed surgically or percutaneously via endoscopy or blind techniques. Enteral feeding may need to be continued for months in some cases. With aggressive nutritional support 75-95% of cats have a good prognosis while without aggressive nutritional support less than 10% of cats will do well. Pancreatitis as a concurrent disease process should be considered in cats not responding to therapy. Refeeding induced hypophosphatemia is a rare complication that can cause hemolysis or neurologic signs that may mimic hepatoencephalopathy. Feline Inflammatory Liver Disease Cholangitis and cholangiohepatitis is a complex of related inflammatory hepatobiliary disorders. They accounted for approximately 26% of the liver diseases reported in cats in one large retrospective study (Gagne, et al. JAVMA, 1999; 214:513). This was second to hepatic lipidosis which accounted for approximately 50% of the cases. Inflammatory liver diseases are characterized by the predominant inflammatory cell infiltrate seen histopathologically. The inflammation is usually seen in the portal areas; and can be characterized as suppurative (neutrophilic), non-suppurative (lymphocytic/plasmacytic); sclerosing lymphocytic cholangitis, or biliary cirrhosis (fibrosis). There have been many terms used in the veterinary literature to describe inflammatory liver diseases prompting the World Small Animal Veterinary Association (WSAVA) Liver Standardization Group to suggest standardized criteria for diagnosis of liver diseases of dogs and cats. The standards define three main forms of cholangitis which are recognized to occur in feline patients: neutrophilic cholangitis, lymphocytic cholangitis, and chronic cholangitis associated with liver fluke infestation. Cholangitis is often associated with periportal necrosis. Neutrophilic cholangitis can be further subdivided into acute (also termed suppurative by some authors) in which neutrophils are seen and chronic in which a mixture of neutrophils and lymphocytes/plasma cells are seen. Lymphocytic cholangitis (formerly lymphocytic portal hepatitis) is the term that has become accepted to describe the histologic

7 classification in which lymphocytes and/or plasma cells are noted to infiltrate the portal areas. This replaces the older term, lymphocytic/plasmacytic cholangiohepatitis. Lymphocytic cholangitis was more common than neutrophilic cholangitis; being seen in 61% of the cats with inflammatory liver disease in the study by Gagne, et al. Although other studies have noted that chronic neutrophilic cholangitis may be more common. Whether these classifications represent different stages in the progression of one disease or are separate etiologic entities is not known. Nor is the underlying etiology of inflammatory liver disease in cats. Bacterial, allergic, and immune mechanisms have all been speculated to be involved. Bacterial cholangitis may either initiate the inflammatory process or perpetuate it early in the disease course. Immune mechanisms probably also play a role especially in chronic neutrophilic cholangitis and lymphocytic cholangitis. Cats with inflammatory hepatobiliary disease, especially those with suppurative disease, may also have pancreatitis and inflammatory bowel disease. The relationship between these three inflammatory conditions is not well worked out but it has been speculated that the underlying initiator of the inflammatory process may affect the liver, the pancreas, and the small intestine concurrently. The term, triaditis has been coined to describe those situations in which inflammation of the liver, pancreas, and small intestine are seen to occur concurrently. While not a very accurate description, the term seems to have stuck. The clinical findings seen in cats with inflammatory liver disease are similar to those seen with hepatic lipidosis and other liver diseases. Vomiting, anorexia, lethargy, and weight loss are typical. Fever is occasionally seen. Diarrhea while not usual is more common than in cats with hepatic lipidosis and may represent the subset of cats with concurrent inflammatory bowel disease. Affected cats are rarely obese. Cats with neutrophilic cholangitis tend to be younger and are more likely to be severely systemically ill when compared to those cats with lymphocytic cholangitis. Any age cat can be affected. Males predominate in populations of cats with neutrophilic cholangitis as compared to those with lymphocytic cholangitis. Suppurative disease often has an acute course while disease characterized by lymphocytic/plasmacytic inflammation may be more chronic. In evaluating liver enzymes, alkaline phosphatase tends not to be as elevated as in cats with hepatic lipidosis and transaminase activities tend to be higher. It is important to note that liver enzymes can be normal, even in cats with significant hepatobiliary inflammation. Neutrophil counts, transaminase activities, and total bilirubin concentrations tend to be higher in cats with neutrophilic cholangitis when compared to cats with lymphocytic cholangitis. All liver enzymes may be normal early in the course of disease, however. Diagnosis is usually dependent on biopsy as FNA is often normal or reveals non-

8 specific changes. Biopsy for both histopathology and culture should be performed if inflammatory liver disease is suspected. The advent of readily available ultrasonography has resulted in Tru-cut needle biopsy becoming the most popular method of obtaining tissue for histopathology. The diagnostic accuracy of Tru-cut obtained biopsies has been questioned (Cole, et al. JAVMA, 2002; 220: ). In the study by Cole, et al. liver biopsies obtained from dogs and cats by tru-cut techniques were compared to wedge biopsies. Paired 18 g Trucut needle biopsies commonly yielded a different diagnosis than wedge biopsy. If it is assumed that the wedge biopsy is the gold standard then the 18 g Tru-cut biopsies were highly inaccurate. Larger samples obtained with a 14 g needle may be more accurate. Laparoscopically obtained samples should be considered when feasible. Prior to biopsy, coagulation parameters should be evaluated. PIVKA may be the most sensitive indicator of potential bleeding tendencies. Vitamin K1 ( mg/kg SQ given within 24 hours of biopsy may decrease the risk of bleeding. In addition to the supportive and nutritional support used to manage cats with hepatic lipidosis, antibiotics should be used when treating cats with inflammatory liver disease. For patients with suspected loss of hepatic function, it is ideal to select drugs that rely on the kidneys for elimination rather than hepatic biotransformation. In the case of antibiotic therapy, the β-lactam antibiotics (penicillin, ampicillin, cephalosporins) are the best choice. (Hepatic reactions observed in people caused by amoxicillin-clavulanate were associated with a specific leukocyte antigen and have not been reported in animals.) The fluoroquinolones (enrofloxacin, marbofloxacin, orbifloxacin, difloxacin) have had a good safety record and increased risk of toxicity in animals with hepatic disease has not been documented. Although some of these drugs are metabolized, the clearance is low and probably not affected unless there is substantial loss of hepatic function. These drugs are also cleared by the kidneys. Fluoroquinolones have been known to cause central nervous system (CNS) problems in susceptible individuals. This is most likely caused by penetration across the blood-brain-barrier (BBB) and inhibition of the action of the GABA neurotransmitter. Problems observed have been seizures, excitement, and disorientation. Animals with seizure disorders caused by hepatic encephalopathy may be more prone to CNS problems caused by fluoroquinolones. If a complication is observed after prescribing a fluoroquinolone drug, a switch to a safer drug is appropriate. Macrolides (erythromycin, azithromycin, and similar drugs) are sometimes used for infections in animals with hepatic disease. There are no specific problems identified in patients with hepatic disease, but these antibiotics are often associated with gastrointestinal problems in

9 animals (diarrhea and vomiting). Therefore, when prescribing these drugs, veterinarians should be careful not to mistake a drug-related problem for an underlying disease, or complicate an already-existing problem. Metronidazole and related drugs (tinidazole, ronidazole) are sometimes used in patients with hepatic disease because of the anaerobic spectrum. They have been safe drugs when prescribed according to standard dose recommendations, but when doses have been exceeded, problems may arise. The most serious problem caused by metronidazole has been attributed to CNS toxicity and include seizures, ataxia, nystagmus, tremors, and rigidity. These signs have been attributed to inferring with the inhibitory neurotransmitter GABA. Because animals with hepatic disease also may be prone to CNS disorders that also share these clinical signs, veterinarians should understand the risks of metronidazole, and become familiar with the signs associated with toxicity. When using other antimicrobials, veterinarians should be aware of the common adverse effects that may occur if a drug accumulates because of a deficiency in metabolism. Drugs to avoid, if possible, include: trimethoprim-sulfonamides, tetracyclines, rifampin, nitrofurantoin, and chloramphenicol. Immunosuppressive agents should be added to the treatment regime in cats with lymphocytic disease and in cats with neutrophilic disease that fail to respond to antibiotics alone or antibiotics and nutraceuticals. Prednisolone (2-4 mg/kg/day initially then slowly tapered to 1 mg/kg QOD) is used most commonly. Other immunosuppressives that may be used in cats responding poorly to glucocorticoids include chlorambucil (1.5 to 4 mg/m 2 twice a week to every other day; approximately 1 mg < 7 lb cat, 2 mg > 7 lb cat) [probably a safer alternative to azathioprine in the cat], azathioprine (0.3 mg/kg q24-72 hrs) [Note that cats are much more sensitive to the myelosuppressive effects of azathioprine than dogs], methotrexate (0.4 mg divided into 3 doses and given over 24 hours and repeated every 7-10 days has been advocated as a pulse therapy but has not been extensively studied). Ursodeoxycholic acid (Actigall ) mg/kg PO SID is a safe treatment alternative that can be used in cats with either suppurative or non-suppurative disease. The drug appears to have multiple actions including shifting the bile acid pool to a less toxic hydrophilic population, a choleretic effect, reducing expression of Class 2 major histocompatibility complex, and an antiinflammatory effect. Vitamin E (aqueous alpha tocopherol, IU/kg/day) has been advocated for its antioxidant effects. SAMe ( mg PO SID; Denosyl NutraMax) is a precursor of glutathione. Glutathione is an important antioxidant that has been shown to be reduced in dogs and cats with liver disease. The nutriceutical SAMe may help replace glutathione. It also may have hepatoprotective effects in preventing programed cell death (apoptosis) that occurs during

10 inflammatory liver disease. Milk thistle (silymarin) is a nutriceutical that is widely used for its hepatoprotective effects. It may be of benefit as an antioxidant, as an antifibrotic agent, or as an aid in hepatic regeneration. Many studies have evaluated its use in people and show mixed results. Studies in dogs and cats are lacking. Anecdotal evidence would suggest it may be useful at a dose of mg/kg PO SID. NutraMax markets a product for cats, Marin, that is a combination of silybin and vitamin E. Silybin is one of the active ingredients in milk thistle. In Marin it is complexed with phosphatidylcholine to increase the bioavailability. The amount of vitamin E in the tablets is lower than is generally recommended in treating liver disease. A combination of silybin and SAMe is available as the nutriceutical Denamarin.

11 A Case-based Approach to the Feline Patient with Liver Disease (Part 2: Inflammatory Liver Disease and Pancreatitis) Joseph Taboada, DVM, Dipl. ACVIM (Small Animal Internal Medicine) School of Veterinary Medicine, Louisiana State University Baton Rouge, Louisiana Pancreatitis is a common inflammatory disease that has historically probably been over diagnosed in dogs and under diagnosed in cats. It can be acute, recurring, or chronic. Acute necrotizing pancreatitis is most common in dogs. It is a pathophysiologic process in which intrapancreatic enzymes are activated resulting in increases in capillary permeability, initiating of vasoactive amines, and direct tissue damage. Vascular injury and tissue necrosis within the pancreas often extends locally to the stomach, duodenum, colon, and liver. Systemic inflammatory response syndrome (SIRS) is a common sequel. Acute necrotizing pancreatitis is seen less commonly than chronic pancreatitis in cats. Chronic pancreatitis is associated with inflammation and fibrosis. Pancreatic stellate cells appear to be important in the pathophysiology of chronic pancreatitis. Stimulated by oxidative stress and cytokines involved in the inflammatory process, activated stellate cells migrate to the periacinar areas to deposit collagen and fibronectin. The fibrosis contributes to obstruction of pancreatic ductules which in turn contributes to inflammation. Fibrosis appears to be important to the pathophysiology of disease in cats. The pathologic characteristics of feline pancreatitis appear to be similar to those in people; especially when compared to that of dogs. Signalment Any age or breed of cat or dog may develop pancreatitis. In the dog middle-aged, obese females are overrepresented. In cats most cases are domestic short-haired cats but Siamese and Persian cats may be overrepresented. Clinical signs The typical presentation in dogs includes sudden onset of vomiting, anorexia, depression, and abdominal pain. There is often a history of recent ingestion of a fatty meal or dietary indiscretion. Presenting clinical signs in cats include anorexia, lethargy, dehydration, hypothermia, and weight loss. Vomiting and abdominal pain are noted less frequently in cats 2017 J.Taboada

12 than in dogs with pancreatitis. Diarrhea is occasionally noted. Dogs and cats with severe pancreatitis may develop ascites and dyspnea associated with pleural effusion. Icterus is a variable finding. An abdominal mass may be noted and was noted in as many as a third of cats in some studies. This is probably representative of only those cats with very severe pancreatitis. Hypotension and shock are severe complications seen commonly in acute necrotizing pancreatitis. While usually idiopathic, pancreatitis has been associated with hyperlipoproteinemia and hypertriglyceridemia, cholinesterase inhibitors such as organophosphates, trauma resulting in hypoperfusion, and drugs such as thiazide diuretics, furosemide, estrogens, azathioprine, L-asparaginase, sulfonamides, tetracycline, metronidazole, H2-receptor blockers, acetaminophen, procainamide, and nitrofurantoin in dogs. In cats toxoplasmosis, FIP, hepatic lipidosis, liver and pancreatic flukes, lymphosarcoma, trauma, fenthion toxicity, idiopathic chylomicronemia, and diabetes mellitus have all been implicated. Potential risk factors for pancreatitis in dogs, such as obesity, dietary indiscretion, a high fat meal, high fat diets, and pre-existing endocrine diseases do not appear to be risk factors in cats. Diagnosis Hematology and serum biochemistry findings are generally non-specific. A mild normocytic, normochromic non-regenerative anemia and leukogram findings consistent with a stress or mild inflammatory leukogram are typical in cats. In dogs a more significant inflammatory leukogram and hemoconcentration would be more likely. A severe neutrophilia with a left shift is usually only seen in severe acute necrotizing pancreatitis. Thrombocytopenia may be noted and is usually mild. Increased ALT, AST, GGT, and alkaline phosphatase activities are typically noted. Bilirubin may be mildly increased but this is not a consistent finding. Dogs are often azotemic but cats are less likely to be azotemic than dogs with pancreatitis. Hyperglycemia is common and may be associated with stress or with the development of diabetes mellitus. The relationship between pancreatitis and diabetes mellitus is well documented in dogs but not well described in cats. However, cats with diabetes mellitus caused by chronic pancreatic inflammation appear to be very sensitive to insulin administration. Anorexia may result in hypokalemia. Hypocalcemia is common but it is rarely severe enough to cause clinical signs. Hypocalcemia may indicate a poor prognosis in cats with pancreatitis. Occasionally hypercalcemia may be noted. Hypoalbuminemia may be noted, especially in dogs and cats with severe disease or concurrent liver disease. Serum cobalamin is low in a high percentage of cats with pancreatitis J.Taboada

13 A lack of sensitive and specific markers of pancreatitis may make antemortem diagnosis of pancreatitis difficult, especially in cats. Serum amylase and lipase have long been used as screening tests for pancreatitis in dogs but are of little use in the diagnosis of pancreatitis in cats. Serum lipase is thought to be the better of the two enzymes for diagnosing pancreatitis in dogs because it is less likely to be increased in renal or other intraabdominal disease. That said, it is neither sensitive nor specific. Feline trypsin-like immunoreactivity (ftli) has been used as a test for pancreatitis in cats but has not proven sensitive. Feline TLI appears to increase acutely but returns to normal very early in the disease course in most cats making it less than ideal as a diagnostic test. It has been suggested that serum concentration of ftli > 100 µg/l is approximately 80-90% specific and 30-60% sensitive for feline pancreatitis. An abnormal result is therefore usually, but not always, associated with pancreatitis. Azotemia may increase ftli and increases have been noted in cats with severe inflammatory bowel disease. An immunoassay for measuring pancreatic lipase [pancreatic lipase immunoreactivity (PLI)] has been developed by the Texas A&M Gastrointestinal Diagnostic Laboratory and licensed to Idexx Laboratories. It is now considered to be the most sensitive and specific test for diagnosing pancreatitis in dogs and cats. Studies in cats with experimental pancreatitis would indicate that fpl increases rapidly after the development of pancreatitis and stays increased for much longer than ftli. In one published study (Forman et al., J Vet Inten Med 2004;18: ), fpl was found to be 80% sensitive for feline pancreatitis, but there was not a significant difference between cats with pancreatitis and healthy cats. The use of fpl is recommended in combination with abdominal ultrasound (see below). Commercial assays for measurement of cpli (Spec cpl ) and fpl (Spec fpl ) are based on the original cpli and fpli technology. The Spec fpl has become available through IDEXX Laboratories. Diagnostic Imaging Radiographs are usually non-specific. Decreased serosal detail may be noted if ascites is present. Decreased detail in the upper right quadrant on the ventrodorsal view may be noted but is seen less commonly in cats than in dogs with pancreatitis. A mass effect may be noted in severe cases. Pleural effusion may be noted on thoracic radiographs. Ultrasound of the pancreas in pancreatitis may reveal a mixed or hypoechogenic pattern, cavitary lesions, dilation of the pancreatic ducts, or evidence of peripancreatic edema and effusion. The pancreas may appear normal on ultrasound in many cats with pancreatitis, but a recent study (Forman et al.) found 80% sensitivity and 88% specificity for ultrasound in diagnosing feline pancreatitis J.Taboada

14 Because of the similar sensitivity and higher specificity of abdominal ultrasound vs. fpli found in this study, ultrasound could be considered the more useful diagnostic test. Further study is needed before firm recommendations can be made. Abdominal computed tomography (CT) has not been shown to be useful in diagnosing pancreatitis in cats. Biopsy The gold standard in diagnosis of pancreatitis is histopathology. Findings in cases of acute pancreatitis include peri-pancreatic fat necrosis, and focal to multifocal pancreatic acinar cell necrosis and inflammation. The inflammation can be quite variable making evaluation of multiple biopsies from different sites critical. Chronic pancreatitis is more common in cats and is usually characterized by variable degrees of fibrosis and lymphocytic inflammation. Fibrosis appears to be more important than inflammation in chronic pancreatitis in cats. The pancreas may appear grossly normal so biopsy is warranted even in cases where the pancreas appears normal at laparotomy or laparoscopy. In a recent study the prevalence of pancreatitis based on histopathology was 67% in cases with GI and other disease and 45% in apparently healthy cats. Treatment (Canine) Depending on the severity of the pancreatitis, treatment of dogs can at times be both difficult and frustrating. In most cases of canine pancreatitis a specific underlying etiology cannot be determined but if it is, specific treatment for that cause should be initiated. Treating shock, rehydration and maintenance of normovolemia are the initial goals of therapy. Fluids such as 0.9% NaCl or lactated Ringers should initially be given. Potassium should be added to the fluids to maintain normal potassium, and glucose should be added if hypoglycemia is noted. Maintaining pancreatic microcirculation may be enhanced by plasma (20ml/kg IV) or colloid (10-20ml/kg/day IV) administration. Plasma transfusion is widely recommended in veterinary medicine but has not been critically evaluated in dogs and cats with pancreatitis. Studies in human patients with pancreatitis have not shown an advantage of plasma therapy. Dextran 70 and hetastarch may have antithrombotic effects that help maintain the microcirculation in addition to their colloidal effects. An external source of heat may be necessary to treat hypothermia, especially in smaller dogs with a larger surface area to weight ratio. Antibiotics should be directed against a bacterial infection if one is suspected. Even if a primary bacterial infection is not suspected, a broad spectrum antibiotic should be given to minimize the effects of bacterial translocation. Abdominal pain is common so analgesics should be used as needed to 2017 J.Taboada

15 keep the patient comfortable. When pain is present, parenteral administration of opioid agonists such as hydromorphone, morphine, and fentanyl provide relief to most patients with severe pain. A fentanyl patch can be applied as an effective means of delivering analgesia in a dog that is likely vomiting. An epidural catheter can be placed for epidural delivery of analgesic medications or local anesthetic can be administered into the caudal thoracic space or cranial peritoneal cavity. Corticosteroids are indicated in acute pancreatitis only if shock is present. There is some controversy as to whether steroids may have negative effects on patients with pancreatitis but there is little convincing evidence that steroids are either helpful or detrimental except in situations such as shock. Immune-mediated pancreatitis, especially in association with systemic lupus erythematosus, is increasingly recognized as an important cause of pancreatitis in people, and steroid therapy has been shown to reduce mortality significantly in these patients. Primary immune-mediated pancreatitis has not been described in dogs and cats, but it should be considered that a definitive cause is not found in most dogs and cats with pancreatitis. Conventional therapeutic approach to patients with pancreatitis would indicate that the patient should be fasted to allow the pancreas to rest. Fasting should result in a physiologic state in which less pancreatic enzyme is being produced and released which may result in reduced pancreatic damage during periods of pancreatic inflammation. The conventional approach is to fast the patient until the clinical signs associated with the pancreatitis have stopped. In some cases the clinical signs may linger for quite some time and nutritional support may become indicated. Dogs are metabolically suited for long fasts and one should not be too quick to start feeding in canine patients with pancreatitis. If nutritional support is deemed necessary, however, ideally, jejunostomy tube feeding or TPN should be considered to reduce pancreatic activity. It should be noted that even when jejunostomy feeding or TPN is used pancreatic activity will be increased over the basal fasted state. The common practice of fasting dogs with pancreatitis is somewhat questionable. The underlying mechanism of pancreatic damage involves abnormally high levels of cholecystokinin, resulting in activation of digestive enzymes within the pancreatic parenchyma and suppression of pancreatic secretion. One could argue that feeding to restore more normal pancreatic secretion might be advantageous. In fact, large-scale studies in human pancreatitis have shown decreased morbidity and mortality when patients are not fasted. Some veterinary internists recommend feeding of dogs with pancreatitis unless vomiting prevents it, and the practice of fasting these patients should probably be reconsidered J.Taboada

16 Treatment (Feline) Georgia Veterinary Medical Association Fall Convention Therapy for cats with pancreatitis is not well described or agreed upon. If a specific underlying etiology is noted it should be treated. Cats with acute necrotizing pancreatitis should be treated with fluids such as 0.9% NaCl or Lactated Ringers. Potassium should be added to the fluids to maintain normal potassium and glucose should be added if hypoglycemia is noted. Treating shock, rehydration and maintenance of normovolemia are the goals. Maintaining pancreatic microcirculation may be enhanced by plasma (20ml/kg IV) or colloid (10-20ml/kg/day IV) administration. Dextran 70 and hetastarch may have antithrombotic effects that help maintain the microcirculation. An external source of heat may be necessary to treat hypothermia. Antibiotics are indicated if bacterial infection or toxoplasmosis is suspected but are not necessary to prevent the bacterial translocation that appears to be common in dogs. Abdominal pain is not common so analgesics are not indicated in most cases. When pain is present, butorphanol ( mg/kg every 4 6 hours) can be given. Buprenorphine ( mg/kg SQ q6-12hrs) or oxymorphone ( mg/kg cats IM, SQ q1-3hrs can also be used but may have a negative effect on respiration. Non-steroidal antiinflammatory agents should probably not be used. Corticosteroids are indicated in acute pancreatitis only if shock is present. Corticosteroids may be indicated in cats with chronic pancreatitis and should be used if concurrent inflammatory bowel disease or liver disease is present. Because cobalamin deficiency is common, supplemental parenteral cobalamin should be considered. Injectable cobalamin can be administered at a dose of 250 ug subcutaneously once a week for 6 weeks, then every 2 weeks for 6 doses, then monthly. Most generic cobalamin preparations contain 1 mg/ml (1000 ug/ml). Most multi-vitamin and B-complex injectable formulations contain significantly lower concentrations of cobalamin. This practice of fasting the patient to rest the pancreas is more controversial in the cat because hepatic lipidosis can be a concurrent problem or a sequela to fasting. Ideally, jejunostomy tube feeding or TPN should be considered but if the patient is not vomiting gastrostomy tube, esophagostomy tube, or nasoesophageal tube feeding are probably appropriately used. Some authors would suggest that treatment indications include use of antioxidants such as vitamin E, SAMe, or Silybin. Supportive studies are lacking but the possible association with inflammatory liver disease and the mechanisms of actions of such nutraceuticals make their use logical. Feline Inflammatory Liver Disease Cholangitis and cholangiohepatitis is a complex of related inflammatory hepatobiliary 2017 J.Taboada

17 disorders. They accounted for approximately 26% of the liver diseases reported in cats in one large retrospective study (Gagne, et al. JAVMA, 1999; 214:513). This was second to hepatic lipidosis which accounted for approximately 50% of the cases. Inflammatory liver diseases are characterized by the predominant inflammatory cell infiltrate seen histopathologically. The inflammation is usually seen in the portal areas; and can be characterized as suppurative (neutrophilic), non-suppurative (lymphocytic/plasmacytic); sclerosing lymphocytic cholangitis, or biliary cirrhosis (fibrosis). There have been many terms used in the veterinary literature to describe inflammatory liver diseases prompting the World Small Animal Veterinary Association (WSAVA) Liver Standardization Group to suggest standardized criteria for diagnosis of liver diseases of dogs and cats. The standards define three main forms of cholangitis which are recognized to occur in feline patients: neutrophilic cholangitis, lymphocytic cholangitis, and chronic cholangitis associated with liver fluke infestation. Cholangitis is often associated with periportal necrosis. Neutrophilic cholangitis can be further subdivided into acute (also termed suppurative by some authors) in which neutrophils are seen and chronic in which a mixture of neutrophils and lymphocytes/plasma cells are seen. Lymphocytic cholangitis (formerly lymphocytic portal hepatitis) is the term that has become accepted to describe the histologic classification in which lymphocytes and/or plasma cells are noted to infiltrate the portal areas. This replaces the older term, lymphocytic/plasmacytic cholangiohepatitis. Lymphocytic cholangitis was more common than neutrophilic cholangitis; being seen in 61% of the cats with inflammatory liver disease in the study by Gagne, et al. Although other studies have noted that chronic neutrophilic cholangitis may be more common. Whether these classifications represent different stages in the progression of one disease or are separate etiologic entities is not known. Nor is the underlying etiology of inflammatory liver disease in cats. Bacterial, allergic, and immune mechanisms have all been speculated to be involved. Bacterial cholangitis may either initiate the inflammatory process or perpetuate it early in the disease course. Immune mechanisms probably also play a role especially in chronic neutrophilic cholangitis and lymphocytic cholangitis. Cats with inflammatory hepatobiliary disease, especially those with suppurative disease, may also have pancreatitis and inflammatory bowel disease. The relationship between these three inflammatory conditions is not well worked out but it has been speculated that the underlying initiator of the inflammatory process may affect the liver, the pancreas, and the small intestine concurrently. The term, triaditis has been coined to describe those situations in which inflammation of the liver, pancreas, and small intestine are seen to occur concurrently. While not a very accurate description, the term seems to have stuck J.Taboada

18 Diagnosis Georgia Veterinary Medical Association Fall Convention The clinical findings seen in cats with inflammatory liver disease are similar to those seen with hepatic lipidosis and other liver diseases. Vomiting, anorexia, lethargy, and weight loss are typical. Fever is occasionally seen. Diarrhea while not usual is more common than in cats with hepatic lipidosis and may represent the subset of cats with concurrent inflammatory bowel disease. Affected cats are rarely obese. Cats with neutrophilic cholangitis tend to be younger and are more likely to be severely systemically ill when compared to those cats with lymphocytic cholangitis. Any age cat can be affected. Males predominate in populations of cats with neutrophilic cholangitis as compared to those with lymphocytic cholangitis. Suppurative disease often has an acute course while disease characterized by lymphocytic/plasmacytic inflammation may be more chronic. In evaluating liver enzymes, alkaline phosphatase tends not to be as elevated as in cats with hepatic lipidosis and transaminase activities tend to be higher. It is important to note that liver enzymes can be normal, even in cats with significant hepatobiliary inflammation. Neutrophil counts, transaminase activities, and total bilirubin concentrations tend to be higher in cats with neutrophilic cholangitis when compared to cats with lymphocytic cholangitis. All liver enzymes may be normal early in the course of disease, however. Diagnosis is usually dependent on biopsy as FNA is often normal or reveals nonspecific changes. Biopsy for both histopathology and culture should be performed if inflammatory liver disease is suspected. The advent of readily available ultrasonography has resulted in Tru-cut needle biopsy becoming the most popular method of obtaining tissue for histopathology. The diagnostic accuracy of Tru-cut obtained biopsies has been questioned (Cole, et al. JAVMA, 2002; 220: ). In the study by Cole, et al. liver biopsies obtained from dogs and cats by tru-cut techniques were compared to wedge biopsies. Paired 18 g Trucut needle biopsies commonly yielded a different diagnosis than wedge biopsy. If it is assumed that the wedge biopsy is the gold standard then the 18 g Tru-cut biopsies were highly inaccurate. Larger samples obtained with a 14 g needle may be more accurate. Laparoscopically obtained samples should be considered when feasible. Prior to biopsy, coagulation parameters should be evaluated. PIVKA may be the most sensitive indicator of potential bleeding tendencies. Vitamin K1 ( mg/kg SQ given within 24 hours of biopsy may decrease the risk of bleeding. Treatment In addition to the supportive and nutritional support used to manage cats with hepatic 2017 J.Taboada

19 lipidosis, antibiotics should be used when treating cats with inflammatory liver disease. For patients with suspected loss of hepatic function, it is ideal to select drugs that rely on the kidneys for elimination rather than hepatic biotransformation. In the case of antibiotic therapy, the β-lactam antibiotics (penicillin, ampicillin, cephalosporins) are the best choice. (Hepatic reactions observed in people caused by amoxicillin-clavulanate were associated with a specific leukocyte antigen and have not been reported in animals.) The fluoroquinolones (enrofloxacin, marbofloxacin, orbifloxacin, difloxacin) have had a good safety record and increased risk of toxicity in animals with hepatic disease has not been documented. Although some of these drugs are metabolized, the clearance is low and probably not affected unless there is substantial loss of hepatic function. These drugs are also cleared by the kidneys. Fluoroquinolones have been known to cause central nervous system (CNS) problems in susceptible individuals. This is most likely caused by penetration across the blood-brain-barrier (BBB) and inhibition of the action of the GABA neurotransmitter. Problems observed have been seizures, excitement, and disorientation. Animals with seizure disorders caused by hepatic encephalopathy may be more prone to CNS problems caused by fluoroquinolones. If a complication is observed after prescribing a fluoroquinolone drug, a switch to a safer drug is appropriate. Macrolides (erythromycin, azithromycin, and similar drugs) are sometimes used for infections in animals with hepatic disease. There are no specific problems identified in patients with hepatic disease, but these antibiotics are often associated with gastrointestinal problems in animals (diarrhea and vomiting). Therefore, when prescribing these drugs, veterinarians should be careful not to mistake a drug-related problem for an underlying disease, or complicate an already-existing problem. Metronidazole and related drugs (tinidazole, ronidazole) are sometimes used in patients with hepatic disease because of the anaerobic spectrum. They have been safe drugs when prescribed according to standard dose recommendations, but when doses have been exceeded, problems may arise. The most serious problem caused by metronidazole has been attributed to CNS toxicity and include seizures, ataxia, nystagmus, tremors, and rigidity. These signs have been attributed to inferring with the inhibitory neurotransmitter GABA. Because animals with hepatic disease also may be prone to CNS disorders that also share these clinical signs, veterinarians should understand the risks of metronidazole, and become familiar with the signs associated with toxicity. When using other antimicrobials, veterinarians should be aware of the common adverse effects that may occur if a drug accumulates because of a deficiency in metabolism. Drugs to avoid, if possible, include: trimethoprim-sulfonamides, tetracyclines, rifampin, nitrofurantoin, and chloramphenicol J.Taboada